Tau undergoes a discrete set of shape changes during filament formation in Alzheimer's disease (AD). Concurrent with these changes are phosphorylation events throughout the molecule and cleavage at a caspase site (D421) that elevates the rate of tau filament formation in vitro. N-terminal truncation events that apparently change the shape of the tau molecule have also been discovered; the timing of their appearance during the course of tau filament formation and maturation in situ is important in understanding the role of modified tau, tau polymers, and tau toxicity in AD neurodegeneration. We hypothesize that such modifications at the carboxy and amino terminal regions of tau will correlate well with the cognitive transition from the non-cognitive impairment (NCI) to mild cognitive impairment (MCI) to AD. We will test this hypothesis in brain regions vulnerable in early AD as follows: 1. We will perform quantitative immunohistochemical analyses using markers of the C-terminus of tau in situ employing brain sections obtained as part of the Religious Orders Study (ROS). Specifically, we will order the events that precede and succeed cleavage of the D421 caspase site using existing monoclonal antibodies to correlate the relationship between C-terminal phosphorylation events and caspase cleavage with individual tests of memory function; 2. Using EM localization and immunochemical studies, we will determine which PHF/SF populations bind to Tau-C3; 3. Using standard protein chemistry coupled with mass spectrometry, we propose to identify N-terminal truncation and adjacent phosphorylation sites in soluble tau and SDS-insoluble PHF-tau. We will then produce antibodies specific for tau cleaved at the most abundant sites; 4. Novel and existing antibodies will be used to stain tissue sections taken from the entorhinal cortices and hippocampi of brains collected by the ROS to determine which of the amino truncations in the tau molecule correlate best with the transition from NCI->MCI->AD; and, 5. NFT staging will be correlated with galanin hyperinnervation and gene expression in Project 3 and with entorhinal cortex and hippocampal volumes in cases from Project 1.